CN103366091B - Based on the abnormal tax return data detection method of multilevel threshold exponent-weighted average - Google Patents
Based on the abnormal tax return data detection method of multilevel threshold exponent-weighted average Download PDFInfo
- Publication number
- CN103366091B CN103366091B CN201310291913.9A CN201310291913A CN103366091B CN 103366091 B CN103366091 B CN 103366091B CN 201310291913 A CN201310291913 A CN 201310291913A CN 103366091 B CN103366091 B CN 103366091B
- Authority
- CN
- China
- Prior art keywords
- vector
- dutiable goods
- data
- tolerance threshold
- statistical indicator
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of abnormal tax return data detection method based on multilevel threshold exponent-weighted average: the history based on taxpayer is declared dutiable goods data, calculate the statistical indicator of data of declaring dutiable goods; Utilize residual index weighting running mean algorithm, the predicted value of each statistical indicator of iterative computation, predicated error and multiple error threshold value; According to multiple error threshold value, detect and extremely declare dutiable goods data and assess exception level.This method effectively can improve the accuracy of detection of abnormal data of declaring dutiable goods, and realizes the assessment of intensity of anomaly.
Description
Technical field:
The present invention relates to Data Detection field, particularly a kind of exception is declared dutiable goods the detection method of data.
Background technology:
Tax audit refers to that the tax authority fulfils obligation to pay tax to taxpayer, withholding agent in accordance with the law, withholds the general name of Tax Check that voluntary situation carries out and work for the treatment of.Tax laws regulation is complicated, audit point is many, and general audit point reaches more than 2000; Simultaneously audit target data are huge, and large enterprise's only financial affairs receipt data one, about have tens million of pen, traditionally manual type, complete one comparatively large enterprises' audit generally need 5-10 people's teamwork 6 months.How to pass through to carry out automatic analysis to the data of declaring dutiable goods of taxpayer, examination goes out abnormal data of declaring dutiable goods, and reduces the data volume of manual audit, becomes one of tax audit field problem demanding prompt solution.
Summary of the invention:
Fundamental purpose of the present invention is to provide a kind of abnormal tax return data detection method based on multilevel threshold exponent-weighted average, adopt the history of multilevel threshold exponent-weighted average Algorithm Analysis taxpayer to declare dutiable goods data, detect whether the data of declaring dutiable goods of taxpayer exist exception.
Object of the present invention is achieved through the following technical solutions:
Based on the abnormal tax return data detection method of multilevel threshold exponent-weighted average, comprise the following steps:
S100, gather the history of taxpayer and to declare dutiable goods data;
S101, the taxpayer's history gathered according to step S100 are declared dutiable goods data, calculate all kinds of statistical indicator of data in each cycle of declaring dutiable goods of declaring dutiable goods;
S102, according to the statistical indicator calculated in step S101, build statistical indicator vector; Declaring dutiable goods the cycle according to statistical indicator, generates the time series S (i) of statistical indicator vector, and S (i) represents the statistical indicator vector in i-th cycle of declaring dutiable goods; EWMA algorithm is adopted to calculate statistical indicator predicted vector PS (i) and the error vector E (i) in i-th cycle of declaring dutiable goods:
PS(i)＝λ
1·S(i)+(1-λ
1)·PS(i-1)
E(i)＝|S(i)-PS(i-1)|
Wherein, smoothing factor λ
1be 0.3, PS (i-1) be the i-th-1 period forecasting vector of declaring dutiable goods, PS (0) is statistical indicator vector S (1) in the 1st cycle of declaring dutiable goods;
S103, employing residual index weighting running mean algorithm, calculate evaluated error vector PE (i) in i-th cycle of declaring dutiable goods:
PE(i)
2＝λ
2·E(i)
2+(1-λ
2)·PE(i-1)
2
Wherein prediction-error coefficients λ
2it is 0.1, PE (0)=0;
Calculate and obtain multistage predicated error tolerance threshold vector;
The error tolerance threshold that S104, contrast are declared dutiable goods in data statistics scale error vector E (i) and multistage predicated error tolerance threshold vector, judges whether respectively to declare dutiable goods cycle every data target containing abnormal count off data.
The present invention further improves and is: the history of taxpayer declare dutiable goods that data comprise all kinds of income, expenditure, each tax category are declared dutiable goods in volume one or more.
The present invention further improves and is: in step S102, the time series generative process of statistical indicator vector is: by the same every statistical indicator Z declared dutiable goods in the cycle
1, Z
2..., Z
n, generate a statistical indicator vector S=(Z
1, Z
2..., Z
n); According to the time sequencing of data of declaring dutiable goods, to the sequence of statistical indicator collection, and generate the time series S (1) of statistical indicator vector, S (2) ..., S (i), wherein S (i)=(Z
1(i), Z
2(i) ..., Z
n(i)), Z
ni () represents n-th statistical indicator in i-th cycle of declaring dutiable goods.
The present invention further improves and is: the method calculating multistage predicated error tolerance threshold vector in step S103 is: arrange the confidence alpha that m level is different
1, α
2..., α
m, calculate the interval value U of normal distribution of corresponding degree of confidence
α 1, U
α 2..., U
α m; Be multiplied with evaluated error vector PE (i), obtain multiple error tolerance threshold vector CL_1, CL_2 ..., CL_m, wherein i cycle xth level error tolerance threshold vector CL_x (i) computing formula of declaring dutiable goods is
CL_x (i)=U
α xpE (i-1); X, m are positive integer, 1≤x≤m.
The present invention further improves and is: in step S104, declares dutiable goods the cycle at i-th, to find in predicated error vector E (i) whether importantly to exceed multistage predicated error tolerance threshold, if containing, the count off data corresponding to this component have exception; If predicated error vector E (i) exceedes multistage predicated error tolerance threshold without any component, then not containing abnormal data in data of declaring dutiable goods.
The present invention further improves and is: in step S104, in the i moment, find in predicated error vector E (i) respective components that whether there is any component and exceed multiple error tolerance threshold vector, if containing, the count off data corresponding to this component have exception, and are divided into different abnormality degrees according to the size of the degree of confidence corresponding to error tolerance threshold vector; If predicated error vector E (i) exceedes multistage predicated error tolerance threshold without any component, then not containing abnormal data in data of declaring dutiable goods.
The present invention further improves and is: multistage predicated error tolerance threshold vector described in step S103 comprises first order error tolerance threshold vector sum second level error tolerance threshold vector; Degree of confidence corresponding to the error tolerance threshold vector of the first order error tolerance threshold vector sum second level is respectively 95% and 99%.
The present invention further improves and is: in step S104, the detection method of abnormal data is: in the i moment, find in predicated error vector prediction error vector E (i) respective components that whether there is any component and exceed multiple error tolerance threshold vector, if there is component E
ji () is greater than the component CL_x of the error tolerance threshold vector of xth level
j(i), and the analysis CL_x+1 being less than (x+1)th level
ji (), then export: " abnormality detection result: Z
j, its abnormality degree is x "; If exceed multistage predicated error tolerance threshold without any component, then export: " no abnormal data ".
Relative to prior art, the invention has the beneficial effects as follows:
(1) algorithm complex is low, is conducive to extensive use: taxpayer's quantity in the whole nation is close to ten million order of magnitude, and the complexity of analytical algorithm directly affects the effect of use; The present invention utilizes residual index Weighted Average Algorithm to analyze taxpayer's historical data, and algorithm complex is low, fast operation, can support the data analysis of extensive taxpayer;
(2) abnormality detection precision is high, realize abnormality degree assessment: conventional method method for detecting abnormality relies on single fiducial interval to set to detect exception, the present invention is by arranging multilevel threshold with the intensity of anomaly of assessment data, on the one hand detection threshold can be set more flexibly, without the need to worrying that threshold value arranges the wrong report that causes and contradiction between failing to report; On the other hand, by assessing the abnormality degree of data, to tax audit, personnel provide decision support, assist its event that notes abnormalities sooner.
The history that the present invention is based on taxpayer is declared dutiable goods data, calculates the statistical indicator of data of declaring dutiable goods; Utilize residual index weighting running mean algorithm, the predicted value of each statistical indicator of iterative computation, predicated error and multiple error threshold value; According to multiple error threshold value, detect and extremely declare dutiable goods data and assess exception level; This method effectively can improve the accuracy of detection of abnormal data of declaring dutiable goods, and realizes the assessment of intensity of anomaly.
Accompanying drawing illustrates:
Fig. 1 is the abnormal tax return data detection method block diagram based on multilevel threshold exponent-weighted average.
Embodiment:
Refer to shown in Fig. 1, a kind of abnormal tax return data detection method based on multilevel threshold exponent-weighted average of the present invention, comprises the following steps:
Step S100, gathers the history of taxpayer and to declare dutiable goods data, comprise all kinds of income, expenditure, each tax category are declared dutiable goods in volume one or more;
Step S101, to declare dutiable goods data according to the history of taxpayer, calculates all kinds of statistical indicator of data in each cycle of declaring dutiable goods of declaring dutiable goods; Select the data of declaring dutiable goods of Shanghai XX enterprise in 2005 to 2012 year as analytic target in this example, select value added tax ratio (A), business tax ratio (B), sales volume year amplification (C) and disbursement year amplification (D) as statistical indicator, concrete numerical value is as shown in table 1;
Table 1 Shanghai XX enterprise in 2005 to 2012 year is declared dutiable goods data statistics index
| 2005 | 2006 | 2007 | 2008 | 2009 | 2010 | 2011 | 2012 | |
| Value added tax ratio | 25.1% | 25.2% | 25.3% | 24.9% | 25.3% | 25.7% | 25.4% | 20.7% |
| Business tax ratio | 14.2% | 13.8% | 13.4% | 12.9% | 13.1% | 12.8% | 12.9% | 21.2% |
| Sell year amplification | 13.5% | 11.2% | 18.6% | 10.5% | 10.6% | 19.1% | 18.8% | 15.3% |
| Expenditure year amplification | 12.1% | 13.3% | 15.9% | 11.2% | 13.8% | 17.4% | 16.9% | 38.1% |
Step S102, according to the statistical indicator calculated in step S101, builds statistical indicator vector S=(value added tax ratio, business tax ratio, sales volume year amplification, disbursement year amplification); According to the time of statistical indicator, generate time series S (the 1)-S (8) of statistical indicator vector, represent the statistical indicator vector of 2005 to 2012 respectively; Adopt EWMA algorithm counting statistics index prediction vector PS (i) and error vector E (i):
PS(i)＝λ
1·S(i)+(1-λ
1)·PS(i-1)
E(i)＝|S(i)-PS(i-1)|
Wherein, smoothing factor λ
1be 0.3, PS (i-1) be upper one predicted vector of declaring dutiable goods the cycle, PS (0) is statistical indicator vector S (1); Counting statistics index prediction vector sum error vector, as shown in table 2 and table 3;
Table 2 is declared dutiable goods data statistics index prediction result
| 2005 | 2006 | 2007 | 2008 | 2009 | 2010 | 2011 | 2012 | |
| Value added tax ratio | 25.1% | 25.1% | 25.2% | 25.1% | 25.2% | 25.3% | 25.3% | 24.0% |
| Business tax ratio | 14.2% | 14.1% | 13.9% | 13.6% | 13.4% | 13.2% | 13.1% | 15.6% |
| Sell year amplification | 13.5% | 12.8% | 14.5% | 13.3% | 12.5% | 14.5% | 15.8% | 15.6% |
| Expenditure year amplification | 12.1% | 12.5% | 13.5% | 12.8% | 13.1% | 14.4% | 15.1% | 22.0% |
Table 3 is declared dutiable goods data statistics scale error result
| 2005 | 2006 | 2007 | 2008 | 2009 | 2010 | 2011 | 2012 | |
| Value added tax ratio | 0 | 0.0010 | 0.0017 | 0.0028 | 0.0020 | 0.0054 | 0.0008 | 0.0464 |
| Business tax ratio | 0 | 0.0040 | 0.0068 | 0.0098 | 0.0048 | 0.0064 | 0.0035 | 0.0806 |
| Sell year amplification | 0 | 0.0230 | 0.0579 | 0.0405 | 0.0273 | 0.0659 | 0.0431 | 0.0048 |
| Expenditure year amplification | 0 | 0.0120 | 0.0344 | 0.0229 | 0.0100 | 0.0430 | 0.0251 | 0.2296 |
Step S103, adopts residual index weighting running mean algorithm, calculates evaluated error vector:
PE(i)
2＝λ
2·E(i)
2+(1-λ
2)·PE(i-1)
2
Wherein prediction-error coefficients λ
2be set to 0.1, calculate evaluated error vector, result of calculation is as shown in table 4;
Table 4 is declared dutiable goods data statistics index evaluated error result
| 2005 | 2006 | 2007 | 2008 | 2009 | 2010 | 2011 | 2012 | |
| Value added tax ratio | 0 | 0.0003 | 0.0006 | 0.0011 | 0.0012 | 0.0021 | 0.0020 | 0.0148 |
| Business tax ratio | 0 | 0.0013 | 0.0025 | 0.0039 | 0.0040 | 0.0043 | 0.0042 | 0.0258 |
| Sell year amplification | 0 | 0.0073 | 0.0196 | 0.0225 | 0.0231 | 0.0302 | 0.0317 | 0.0301 |
| Expenditure year amplification | 0 | 0.0038 | 0.0115 | 0.0131 | 0.0128 | 0.0182 | 0.0190 | 0.0748 |
set 2 grades of error tolerance threshold, degree of confidence is respectively 95% and 99%, calculates the interval value U of normal distribution of degree of confidence
95%=1.96 and U
99%=2.58, calculate 2 grades of error tolerance threshold vectors thus, as shown in table 5 and table 6;
Table 5 the 1st grade of error tolerance threshold (95% fiducial interval)
| 2005 | 2006 | 2007 | 2008 | 2009 | 2010 | 2011 | 2012 | |
| Value added tax ratio | 0 | 0.0006 | 0.0012 | 0.0021 | 0.0023 | 0.0040 | 0.0039 | 0.0290 |
| Business tax ratio | 0 | 0.0025 | 0.0048 | 0.0076 | 0.0078 | 0.0084 | 0.0082 | 0.0505 |
| Sell year amplification | 0 | 0.0143 | 0.0384 | 0.0442 | 0.0452 | 0.0592 | 0.0622 | 0.0591 |
| Expenditure year amplification | 0 | 0.0074 | 0.0225 | 0.0256 | 0.0251 | 0.0357 | 0.0373 | 0.1466 |
Table 6 the 2nd grade of error tolerance threshold (99% fiducial interval)
| 2005 | 2006 | 2007 | 2008 | 2009 | 2010 | 2011 | 2012 | |
| Value added tax ratio | 0 | 0.0008 | 0.0016 | 0.0027 | 0.0031 | 0.0053 | 0.0051 | 0.0382 |
| Business tax ratio | 0 | 0.0033 | 0.0064 | 0.0100 | 0.0103 | 0.0110 | 0.0108 | 0.0665 |
| Sell year amplification | 0 | 0.0188 | 0.0505 | 0.0582 | 0.0595 | 0.0780 | 0.0819 | 0.0778 |
| Expenditure year amplification | 0 | 0.0098 | 0.0296 | 0.0337 | 0.0330 | 0.0470 | 0.0491 | 0.1930 |
Step S104, contrast table 3 is declared dutiable goods the error tolerance threshold in data statistics scale error result and table 5 and table 6, obtains the abnormality degree of every data target of each year, judges whether respectively to declare dutiable goods cycle every data target containing abnormal count off data; Result is as shown in table 7;
Table 7 is declared dutiable goods, and (NULL represents (data statistics scale error of declaring dutiable goods result is less than corresponding all error tolerance threshold) without exception data exception degree assessment result, II represents 2 grades of exceptions (data statistics scale error of declaring dutiable goods result is less than corresponding all error tolerance threshold and is greater than 1 grade and 2 grades of error tolerance threshold), I represents 1 grade of exception (data statistics scale error of declaring dutiable goods result is less than corresponding all error tolerance threshold and is greater than 1 grade of error tolerance threshold, is less than 2 grades of error tolerance threshold))
Table 7 is declared dutiable goods data exception degree assessment result
| 2005 | 2006 | 2007 | 2008 | 2009 | 2010 | 2011 | 2012 | |
| Value added tax ratio | NULL | II | II | II | NULL | II | NULL | II |
| Business tax ratio | NULL | II | II | NULL | NULL | NULL | NULL | II |
| Sell year amplification | NULL | II | II | NULL | NULL | II | NULL | NULL |
| Expenditure year amplification | NULL | II | II | NULL | NULL | II | NULL | II |
Analysis result can be found out, within 2006,2007,2010 and 2012, existing significantly abnormal, finding by analyzing, and the exception of 2006 and 2007 causes because model is in the data training stage; The exception of 2010 is that economic environment in 2010 gets warm again after a cold spell, and causing appears significantly increasing in enterprise's indices; The value added tax of the abnormal show enterprise of 2012 declines to a great extent and business tax increases substantially, and may be that taxpayer transforms the tax category in violation of rules and regulations and causes, the amplification of expenditure cost also exists remarkable exception simultaneously.Therefore, the data of declaring dutiable goods exporting enterprise in 2012 exist abnormal, and wherein value added tax ratio, business tax ratio and expenditure year amplification data exist significantly abnormal.
Claims (1)
1. based on the abnormal tax return data detection method of multilevel threshold exponent-weighted average, it is characterized in that, comprise the following steps:
S100, gather the history of taxpayer and to declare dutiable goods data; Described history declare dutiable goods that data comprise all kinds of income, expenditure, each tax category are declared dutiable goods in volume one or more;
S101, the taxpayer's history gathered according to step S100 are declared dutiable goods data, calculate all kinds of statistical indicator of data in each cycle of declaring dutiable goods of declaring dutiable goods; Described statistical indicator is one or more in value added tax ratio (A), business tax ratio (B), sales volume year amplification (C) and disbursement year amplification (D), is expressed as Z
1, Z
2..., Z
n;
S102, according to the statistical indicator calculated in step S101, build statistical indicator vector; Declaring dutiable goods the cycle according to statistical indicator, generates the time series S (i) of statistical indicator vector, and S (i) represents the statistical indicator vector in i-th cycle of declaring dutiable goods; EWMA algorithm is adopted to calculate statistical indicator predicted vector PS (i) and the error vector E (i) in i-th cycle of declaring dutiable goods:
PS(i)＝λ
1·S(i)+(1-λ
1)·PS(i-1)
E(i)＝|S(i)-PS(i-1)|
Wherein, smoothing factor λ
1be 0.3, PS (i-1) be the i-th-1 period forecasting vector of declaring dutiable goods, PS (0) is statistical indicator vector S (1) in the 1st cycle of declaring dutiable goods;
The time series generative process of statistical indicator vector is: by the same every statistical indicator Z declared dutiable goods in the cycle
1, Z
2..., Z
n, generate a statistical indicator vector S=(Z
1, Z
2..., Z
n); According to the time sequencing of data of declaring dutiable goods, to the sequence of statistical indicator collection, and generate the time series S (1) of statistical indicator vector, S (2) ..., S (i), wherein S (i)=(Z
1(i), Z
2(i) ..., Z
n(i)), Z
ni () represents n-th statistical indicator in i-th cycle of declaring dutiable goods;
S103, employing residual index weighting running mean algorithm, calculate evaluated error vector PE (i) in i-th cycle of declaring dutiable goods:
PE(i)
2＝λ
2·E(i)
2+(1-λ
2)·PE(i-1)
2
Wherein prediction-error coefficients λ
2it is 0.1, PE (0)=0;
Calculate and obtain multiple error tolerance threshold vector;
The method calculating multistage predicated error tolerance threshold vector is: arrange the confidence alpha that m level is different
1, α
2..., α
m, calculate the interval value U of normal distribution of corresponding degree of confidence
α 1, U
α 2..., U
α m; Be multiplied with evaluated error vector PE (i), obtain multiple error tolerance threshold vector CL_1, CL_2 ..., CL_m, wherein i cycle xth level error tolerance threshold vector CL_x (i) computing formula of declaring dutiable goods is:
CL_x (i)=U
α xpE (i-1); X, m are positive integer, 1≤x≤m
S104, compare error vector E (i) and the error tolerance threshold in multiple error tolerance threshold vector, judge whether respectively to declare dutiable goods cycle every data target containing abnormal count off data;
In step S104, declare dutiable goods the cycle at i-th, to find in error vector E (i) and whether importantly exceed multiple error tolerance threshold, if containing, the count off data corresponding to this component have exception; If error vector E (i) exceedes multiple error tolerance threshold without any component, then not containing abnormal data in data of declaring dutiable goods;
In step S104, in the i moment, find in error vector E (i) respective components that whether there is any component and exceed multiple error tolerance threshold vector, if containing, the count off data corresponding to this component have exception, and are divided into different abnormality degrees according to the size of the degree of confidence corresponding to error tolerance threshold vector; If error vector E (i) exceedes multiple error tolerance threshold without any component, then not containing abnormal data in data of declaring dutiable goods;
The tolerance threshold of multiple error described in step S103 vector comprises first order error tolerance threshold vector sum second level error tolerance threshold vector; Degree of confidence corresponding to the error tolerance threshold vector of the first order error tolerance threshold vector sum second level is respectively 95% and 99%.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310291913.9A CN103366091B (en) | 2013-07-11 | 2013-07-11 | Based on the abnormal tax return data detection method of multilevel threshold exponent-weighted average |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310291913.9A CN103366091B (en) | 2013-07-11 | 2013-07-11 | Based on the abnormal tax return data detection method of multilevel threshold exponent-weighted average |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103366091A CN103366091A (en) | 2013-10-23 |
| CN103366091B true CN103366091B (en) | 2015-08-26 |
Family
ID=49367418
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310291913.9A Active CN103366091B (en) | 2013-07-11 | 2013-07-11 | Based on the abnormal tax return data detection method of multilevel threshold exponent-weighted average |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103366091B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104102833B (en) * | 2014-07-10 | 2015-08-05 | 西安交通大学 | Based on the tax index normalization found between compact district and fusion calculation method |
| CN104517232B (en) * | 2014-12-19 | 2018-07-17 | 西安交通大学 | A method of excavating the association taxpayer group that taxable amount is uprushed |
| CN108880841A (en) * | 2017-05-11 | 2018-11-23 | 上海宏时数据系统有限公司 | A kind of threshold values setting, abnormality detection system and the method for service monitoring system |
| CN107086944B (en) * | 2017-06-22 | 2020-04-21 | 北京奇艺世纪科技有限公司 | Anomaly detection method and device |
| CN108667686B (en) * | 2018-04-11 | 2021-10-22 | 国电南瑞科技股份有限公司 | Credibility evaluation method for network message time delay measurement |
| CN111325472A (en) * | 2020-02-28 | 2020-06-23 | 北京思特奇信息技术股份有限公司 | Abnormal data detection method and system |
| CN114445207B (en) * | 2022-04-11 | 2022-07-26 | 广东企数标普科技有限公司 | Tax administration system based on digital RMB |
| CN115801901B (en) * | 2023-01-05 | 2023-08-04 | 安徽皖欣环境科技有限公司 | Enterprise production emission data compression processing method |
| CN116342301B (en) * | 2023-03-08 | 2023-11-28 | 深圳欧税通技术有限公司 | Cross-border enterprise tax declaration condition monitoring and management system based on big data |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2002039254A1 (en) * | 2000-11-09 | 2002-05-16 | Spss Inc. | System and method for building a time series model |
| CN101615186A (en) * | 2009-07-28 | 2009-12-30 | 东北大学 | A kind of BBS user's abnormal behaviour auditing method based on Hidden Markov theory |
| CN101916335A (en) * | 2010-08-19 | 2010-12-15 | 河北农业大学 | Prediction method of city water-requirement time series-exponent smoothing model |
| CN102539823A (en) * | 2012-01-13 | 2012-07-04 | 重庆大学 | Method for forecasting wind speed distribution of WTG (wind turbine generator) |
-
2013
- 2013-07-11 CN CN201310291913.9A patent/CN103366091B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2002039254A1 (en) * | 2000-11-09 | 2002-05-16 | Spss Inc. | System and method for building a time series model |
| CN101615186A (en) * | 2009-07-28 | 2009-12-30 | 东北大学 | A kind of BBS user's abnormal behaviour auditing method based on Hidden Markov theory |
| CN101916335A (en) * | 2010-08-19 | 2010-12-15 | 河北农业大学 | Prediction method of city water-requirement time series-exponent smoothing model |
| CN102539823A (en) * | 2012-01-13 | 2012-07-04 | 重庆大学 | Method for forecasting wind speed distribution of WTG (wind turbine generator) |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103366091A (en) | 2013-10-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103366091B (en) | Based on the abnormal tax return data detection method of multilevel threshold exponent-weighted average | |
| Tinoco et al. | Financial distress and bankruptcy prediction among listed companies using accounting, market and macroeconomic variables | |
| CN103377454B (en) | Based on the abnormal tax return data detection method of cosine similarity | |
| Chen et al. | Detecting biotechnology industry's earnings management using Bayesian network, principal component analysis, back propagation neural network, and decision tree | |
| CN104881783A (en) | E-bank account fraudulent conduct and risk detecting method and system | |
| Tan et al. | A wavelet based investigation of long memory in stock returns | |
| CN106447075A (en) | Industry electricity utilization demand prediction method and industry electricity utilization demand prediction system | |
| CN112016843A (en) | Organizational finance and tax data risk analysis method and related device | |
| Kolbaşi et al. | A comparison of the outlier detecting methods: an application on Turkish foreign trade data | |
| Kozmenko et al. | Statistical model of risk assessment of insurance companys functioning | |
| CN112329862A (en) | Decision tree-based anti-money laundering method and system | |
| Smithson et al. | Quantifying operational risk | |
| CN109887253B (en) | Correlation analysis method for petrochemical device alarm | |
| CN110930258B (en) | Accounts receivable financing variable scale prediction method and system | |
| Eslamloueyan et al. | The effects of the COVID-19 pandemic, economic sanctions, and institutional quality on the poverty gap: The case of Iran | |
| CN112465397A (en) | Audit data analysis method and device | |
| CN113835947A (en) | Method and system for determining abnormality reason based on abnormality identification result | |
| Soofi et al. | Testing for long memory in the Asian foreign exchange rates | |
| Sintiya et al. | SARIMA and Holt-Winters Seasonal Methods for Time Series Forecasting in Tuberculosis Case | |
| Xiangyu et al. | Intelligent identification of corporate tax evasion based on LM neural network | |
| Nenkova et al. | Government expenditure efficiency and macroeconomic performance of Balkan countries: DEA approach | |
| Liu et al. | Crowding in or crowding out? The effect of imported environmentally sound technologies on indigenous green innovation | |
| Putra et al. | Analysis the Effect of Financial Performance Ratios on Profitability at PT. Bank Central Asia Tbk (BCA) 2018-2022 | |
| US20230394069A1 (en) | Method and apparatus for measuring material risk in a data set | |
| Subhan et al. | THE INFLUENCE OF ECONOMIC FACTORS ON THE STOCK PRICE OF KIMIA FARMA COMPANIES ON THE INDONESIAN STOCK EXCHANGE |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C41 | Transfer of patent application or patent right or utility model | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20160419 Address after: 310053, tax building, No. 3738 South Ring Road, Hangzhou, Zhejiang, Binjiang District Patentee after: Servyou Software Group Co., Ltd. Address before: 710049 Xianning West Road, Shaanxi, China, No. 28, No. Patentee before: Xi'an Jiaotong University |